Influence of colloidal fouling and feed water recovery on salt rejection of RO and NF membranes

The influence of colloidal fouling and feed water recovery (or concentration factor, CF) on salt rejection of thin-film composite reverse osmosis (RO) and nanofiltration (NF) membranes was investigated. Fouling experiments were carried out using a laboratory-scale crossflow test unit with continuous permeate disposal to simulate the CF and recovery as commonly observed in full-scale RO/NF systems. For feed waters containing only salt (NaCl), permeate flux declined linearly as CF was increased and salt rejection was nearly constant for both RO and NF membranes. On the other hand, a sharp decrease in permeate flux and significant decline in salt rejection with increasing CF were observed under conditions where colloidal fouling takes place. For both RO and NF membranes, the marked permeate flux decline was attributed to the so-called “cake-enhanced osmotic pressure”. The decline in salt rejection when colloidal fouling predominated was much more substantial for NF than for RO membranes. In all cases, the decline in salt rejection was higher under conditions of more severe colloidal fouling, namely at higher ionic strength and initial permeate flux.     

Investigations of the coupled effect of cake-enhanced osmotic pressure and colloidal fouling in RO using crossflow sampler-modified fouling index ultrafiltration

The aim of this study was to investigate the potential of using the resistivity and deposition rate data from Crossflow Sampler-Modified Fouling Index Ultrafiltration (CFS-MFI_UF) measurements to determine the coupled effects of colloidal fouling and cake enhanced osmotic pressure (CEOP) effect. Cake filtration derived from CFS-MFI_UF was combined with a CEOP model to predict the crossflow RO fouling profile under constant flux filtration. The prediction based on resistivity, I′ from CFS-MFI_UF measurement alone was found to underestimate the RO fouling for high salinity solutions. However, when incorporating the mass information from the CFS-MFI_UF test to account for the CEOP effect, the prediction showed good agreement with the TMP profile of the RO system. The results indicated that the CFS-MFI_UF test which includes the CEOP effect is a very promising technique to provide an estimation of the RO colloidal fouling profile. When the changes of cake thickness and porosity throughout the filtration were considered, the predicted TMP profile based on the model clearly indicated a two-stage of fouling profile which agreed well with the experimental data. Additional studies on the effects of cake thickness and porosity on CEOP highlighted the important influence of cake structure on CEOP.     

The effect of naturally occurring biopolymers on polyamide membrane fouling during surface water treatment

Parallel experiments using a blend of surface waters were conducted to evaluate differential fouling rates among reverse osmosis (RO) membranes when operated under pilot- vs. full-scale conditions. Testing was conducted using a 230 L/min conventional (rapid mix/flocculation/sedimentation/filtration) package plant (CPP) and a 2,000 ML/d fullscale treatment plant (FTP) as pretreatment to separate RO membrane test units. Coagulation consisted of 10 mg/L alum (as Al₂(SO₄)₃-14H₂O) and 2.0 mg/L cationic polymer. A2.5-3.0 mg/L free-chlorine residual was maintained at the filter effluent and converted to chloramines through ammonium sulfate addition (3:1 chlorine-to-ammonia w/w ratio). Membrane performance was based on normalized flux and salt rejection data. Membrane surface analyses included scanning electron microscopy, energy-dispersive spectroscopy, and attenuated total reflectance Fourier transform infrared spectroscopy. Microbial activity and community analyses were conducted through (a) fluorescence staining with 4′,6′-diamidino-2-phenylindole, (b) polymerase-chain reaction amplification of isolated bacterial DNA, and (c) microscopic taxonomic identification. Results indicated that the RO membrane fed by the CPP fouled at least three times faster than the RO membrane fed by the FTP. The differential fouling between the two process streams was determined to be from lack of maintenance in the CPP influent piping that led to the establishment of biological communities consisting of algae, microbes, and, potentially, freshwater clams. These communities produced low levels of natural polymers, which when presented to the polyamide RO membrane surface, resulted in rapid fouling.     

Modeling of RO Flux Decline in Textile Wastewater Reclamation Plants Using Variable Fouling Index

Abstract

This research proposes the model for estimation of flux decline using the concept of “variable fouling index.” The change in fouling index is induced by the accumulative foulant on the membrane and by the adhesion of the foulant on the membrane. The analysis of operating data of textile wastewater reclamation plants using the variable fouling index concept has shown that this model which was derived from short-term data of bench scale achieves a promising accuracy for the estimation of flux decline and increase of net driving pressure in a long-term operation of pilot-scale and full-scale system.     

Mechanisms underlying the effects of membrane fouling on the nanofiltration of trace organic contaminants

The influence of membrane fouling on the retention of four trace organic contaminants - namely sulfamethoxazole, ibuprofen, carbamazepine, and triclosan - by nanofiltration membranes was investigated in this study. Humic acid, alginate, bovine serum albumin, and silica colloids were selected as model foulants to simulate various organic fractions and colloidal matter that are found in secondary treated effluent and surface water. The effects of membrane fouling on the separation process was delineated by comparing retention values of clean and fouled membranes and relate them to the membrane properties (under both clean and fouled conditions) as well as physicochemical characteristics of the trace organic contaminants. Membrane fouling was dependent on the physicochemical properties of the model foulants. Initial foulant-membrane interaction could probably be a major factor governing the process of membrane fouling particularly by the organic foulants. Such membrane-foulant interaction was also a dominating factor governing the effects of membrane fouling on the membrane separation efficacy. In good agreement with our previous study (Nghiem and Hawkes, 2007 [1]), the effects of fouling on retention were found to be membrane pore size dependent. In addition, results reported here suggest that these effects could also be foulant dependent. It was probable that the influence of membrane fouling on trace organic retention could be governed by four distinctive mechanisms: modification of the membrane charge surface, pore blocking, cake enhanced concentration polarisation, and modification of the membrane hydrophobicity. The presence of the fouling layer could affect the retention behavior of charged solutes by altering the membrane surface charge density. While the effect of surface charge modification was clear for inorganic salts, it was less obvious for the negatively charged pharmaceutical species (sulfamethoxazole and ibuprofen) examined in this investigation, possibly due to the interference of the pore blocking mechanism. Evidence of the cake enhanced concentration polarisation effect was quite clear, particularly under colloidal fouling conditions. In addition, organic fouling could also interfere with the solute-membrane interaction, and therefore, exerted considerable influence on the separation process of the hydrophobic trace organic contaminant triclosan.     

反渗透/纳滤膜剖检分析与膜污染诊断研究进展

反渗透（RO）/纳滤（NF）膜元件在长期运行过程中会不可避免地发生膜污染,当产水水质无法满足应用指标时,就需要对膜元件进行更换。膜剖检分析是研究和确定膜污染最直观有效的方法,通过膜剖检分析及膜污染诊断可以为膜元件的日常维护、膜系统运行优化和膜性能修复提供有效依据。但是,目前对于膜剖检分析的实践及膜污染诊断研究还不系统、不全面。本文针对RO/NF膜剖检分析及膜污染诊断相关研究,介绍了膜元件剖检分析流程和各类膜污染分析方法,分析了实际应用中存在的问题,并根据膜剖检分析的意义和价值,重点综述了膜污染成分诊断、膜污染分布情况诊断、不同应用场景膜污染情况对比和不同膜材料的污染情况对比研究进展,以期为膜污染机制研究、膜污染预防控制和膜系统运行改善提供参考。     

Effect of cake layer structure on colloidal fouling in reverse osmosis membranes

A series of reverse osmosis (RO) membrane filtration experiments was performed systematically in order to investigate the effects of various hydrodynamic and physicochemical operational parameters on a cake layer formation in colloidal and particulate suspensions. Bench-scale fouling experiments with a thin-film composite RO membrane were performed at various combinations of trans-membrane pressure (TMP), cross-flow velocity (CFV), particle size, pH, and ionic strength. In this study, silica particles with two different mean diameters of 0.1 and 3.0 μm were used as model colloids. Membrane filtration experiments with colloidal suspensions under various hydrodynamic operating conditions resulted that more significant permeate flux decline was observed as TMP increased and CFV decreased, which was attributed to the higher accumulative mass of particles on the membrane surface. Results of fouling experiments under various physicochemical operating conditions demonstrated that the rate of flux decline decreased significantly with an increase of the ionic strength as well as particle size, while the flux decline rate did not vary when solution pH changed. The experimentally measured cake layer thickness increased with a decrease in particle size and solution ionic strength. Furthermore, the model estimation of cake layer thickness by using a cake filtration theory based on the hydraulic resistance of membrane and cake layer was performed under various ionic strength conditions. The primary model parameters including accumulated mass and specific cake resistance were calculated from the cake layer resistance. This result indicated that the formation of cake layer could be closely related with solution water chemistry. The model estimated cake layer thickness values were in good agreement with the experimentally measured values.     

Laboratory RO and NF processes fouling investigation by residence time distribution curves examination

The fouling phenomena of nanofiltration of calcium sulfate saturated feed solution in the presence of magnesium and carbonates as well as the reverse osmosis of saturated silica feedwater with magnesium chloride content at different pH levels, was investigated by analyzing the membrane residence time distribution curves (RTD) in our laboratory. The fraction of degraded membrane (parameter a), the fraction of active membrane area (parameter b) and the fraction of membrane blocked by impermeable layer (parameter c) as a function of permeate recovery was determined. It was found that not only surface blockage affects permeate efficiency, but also other factors, such as module retentate chamber pressure drop and osmotic pressure increase. The membranes RTD measurements as well as fouling, a, b and c parameters, gathered with ion-selective electrodes (chloride, calcium and cupric ion-selective) and appropriate tracer solutions (sodium chloride, calcium chloride and cupric nitrate) were compared with commonly used conductivity sensor and sodium chloride tracer solution. We found that both the tracer and detector type strongly affects membrane mean residence time and its variance, but at the same time no influence of detector and tracer type on membrane fouling a, b and c parameters was observed.     

炼化污水抗污染反渗透膜回用处理中试研究

采用氧化混凝-复合过滤-反渗透组合工艺处理大庆炼化公司外排污水，考察了预处理系统的操作参数和抗污染反渗透膜组件性能的变化规律，为工程设计提供了基本参数。结果表明，预处理出水水质主要受过滤速度和絮凝剂加药量的影响，而膜污染是影响反渗透系统运行的主要因素。系统出水水质符合工业用水软化除盐设计规范。     

Chemical control of colloidal fouling of reverse osmosis systems

Fouling of membranes by colloidal organic and inorganic particles continues to be documented as the most common and challenging obstacle in attaining stable continuous operation of reverse osmosis (RO) and ultrafiltration (UF) systems. Much current research is being conducted on physical parameters to mitigate such fouling. The focus has been on membrane synthesis and element design; microfiltration and ultrafiltration pretreatment; electromagnetic devices; correlation with physical factors such as Silt Density Index, zeta potential and critical flux; technique of direct observation of fouling process through a membrane; and classification of macromolecular organics for correlation with fouling characteristics. We report initial successes with chemical control of colloidal fouling. Through screening with a large number of observable coagulations of natural colloids, we have developed a group of proprietary anticoagulants and dispersants that would, at less than 10 ppm dosage to the RO feedwater, control various classes of colloidal foulants. Case studies of the control of humic matter, elemental sulfur and colloidal silicate in problematic RO systems that became stabilized are briefly presented. We conclude that a great need and potential exists in economically controlling the myriads of fouling interactions of colloidal particles during concentration within the brine channels of RO membrane elements. Low dosages of antifoulants can in many cases obviate the need for installation and maintenance of pretreatment unit or operations designed to remove such colloidal foulants from the process stream.     

Synergistic behavior between silica and alginate: Novel approach for removing silica scale from RO membranes

The formation of mineral scale deposits on membranes is a pervasive and expensive problem for the water treatment industry. A series of experiments run on a laboratory-scale reverse osmosis membrane system examined the fouling of membranes when the feed water was spiked with organic and inorganic foulants. Alginic acid was used as the organic foulant and silica was used as the inorganic foulant. Studies involving interactions of these two foulants have not previously been reported in literature. Experiments were run with each foulant individually to characterize fouling at different velocities and pressures. Experiments were then run using both foulants together to characterize the synergistic effects on membrane fouling. One set of experiments with both foulants demonstrated that alginic acid inhibits silica fouling on reverse osmosis membranes. Further experiments indicated that alginic acid added after silica fouling had already occurred was able to remove silica scale from the membrane and restore permeate flux.     

Organic fouling of RO membranes: Investigating the correlation of RO and UF fouling resistances for predictive purposes

This study examines the characteristics of deposits from dead-end UF and cross-flow RO filtration of dilute mixtures of sodium alginate (SA) and humic acids (HA), which are representative of common foulants in RO and NF feed waters. The data are interpreted in terms of specific cake resistance, α. The presence of alginates in the mixed fouling layers, even in relatively small percentage, imparts characteristic properties to these deposits; i.e. a gel-like structure which is associated with reduced compressibility and permeability. With SA/HA mixtures, high rejection of organic species and high deposition factors are obtained in UF and RO tests, respectively. The specific resistance of mixed SA and HA deposits tends to increase with increasing SA concentration in solution. The resistances α exhibit fairly strong (power law) dependence on the pressure difference across the cake, ΔΡ_c, for both RO and UF tests. The fairly satisfactory correlation of resistance α versus ΔΡ_c data, obtained from both types of tests (for the mixtures of organic species tested at small salinity) is a significant result of this work. Based on this correlation, an approach is suggested for estimating fouling index I and initial membrane fouling rate of a specific RO process, relying on limited UF tests.     

RO and NF membrane fouling and cleaning and pore size distribution variations

The variations of porosity parameters of some reverse osmosis (RO) and nanofiltration (NF) polyamide thin-film composite membranes were determined in order to explain the changes of membranes' performances caused by membrane fouling and chemical cleaning of the fouled membranes. The pore size distribution curves and the effective number of pores in the membrane surface indicated plugging of the tight network pores in the membrane surface and even their disappearance during fouling. The enlargement of the wider aggregate pores was responsible for the noticed reduction in salt rejection. The initial pore structure of the fouled RO membrane was restored by immediate chemical cleaning. A delay of chemical cleaning of the fouled membranes led to irreversible changes in the porous structure of both the RO and NF membranes, which were caused by a microbial activity.     

Humic acid fouling in the membrane distillation process

This work investigates the extent of humic acid fouling during the membrane distillation process for water treatment. The effects of pH, ionic strength, and divalent ion on fouling were studied. The experiments were performed with a 0.22-μm PVDF flat-sheet membrane in a direct contact membrane distillation unit. Flux declines were negligible (less than 6%) for the ranges of humic acid concentration, ionic strength, and pH studied. The examination of the membrane surface by SEM revealed a thin deposit layer. The addition of divalent cations (Ca²⁺) into the solution considerably reduced flux when Ca²⁺ concentration exceeded the critical coagulation concentration. Ca²⁺ affected flux by forming complexes with humic acids and resulted in coagulation on the membrane surface. The normalized flux, J/J₀, was 0.57 after 18 h of operation when the CaCl₂ concentration was 3.775 mM. However, the deposit of humic acid coagulate on the membrane surface was loosely packed, and was rather easily removed. Rinsing of the fouled membrane with clean water and a 0.1 M NaOH solution gave 100% of flux recovery.     

Performances of RO and NF processes for wastewater reuse: Tertiary treatment after a conventional activated sludge or a membrane bioreactor

Wastewater reclamation requires processes and technologies having the ability to reduce the presence of micropollutants which are not wholly treated in conventional WWTP. Due to the complexity of membrane-solute interactions and the diversity of secondary treatment effluent (STE) matrices, deeper investigations are required to identify the major foulant species and more specifically their behaviour at high concentration in real waters. This study investigates the rejection and fouling potential of nanofiltration (NF) and low-pressure reverse osmosis (RO) membranes with two STEs sampled from i) a conventional activated sludge process coupled with ultrafiltration (CAS-UF) and from ii) a membrane bioreactor MBR (AquaRM^®, SAUR (France)). Whatever the origin of the effluent, RO seems to be the best solution to prevent pollution of tertiary effluents (expected result) but also to obtain low fouling levels. The different composition and molecular weight distribution of MBR and CAS-UF effluents can explain the different fouling behaviours that were observed.     

Effects of retained natural organic matter (NOM) on NOM rejection and membrane flux decline with nanofiltration and ultrafiltration

Two natural source waters containing natural organic matter (NOM) with different physical and chemical characteristics were dead-end filtered using five types of membranes having different material and geometric properties. In this study retained dissolved organic carbon (DOC) per unit membrane area is introduced as a better parameter compared to permeate volume, time, and delivered DOC to provide a reasonable comparison of NOM rejection and flux-decline trends. Retained DOC/NOM was calculated, which influences NOM concentration polarization at the membrane interface, and transport measurements of NOM rejection and flux decline were made. Molecular weight (MW) distribution measurements (by size exclusion chromatography) were used to calculate the average MW of the NOM. This persuasively demonstrated that the nominal molecular weight cut-off (MWCO) of a membrane is not the unique predictor of rejection characteristics for NOM composites. The charge density of NOM from the source waters was measured to estimate its effects on NOM rejection and flux decline during filtration. The contact angle of the membranes was used to determine hydrophobic interactions between NOM and membrane. All filtration measurements were performed at approximately the same permeate flow rate in order to minimize artifacts from mass transfer at the membrane interface. ESNA having a nominal MWCO of 200 Daltons showed NOM rejection greater than 95% and flux decline lower than 10% under a condition of a retained DOC of 0.5 mg C/cm² for the feed source waters. The other membranes having larger membrane pores (nominal MWCOs ranging from 8,000 to 20,000 Daltons) than the ESNA showed NOM rejection ranging from 68% to 86% and flux decline ranging from 5% to 17% at the same retained DOC for the waters.     

On the fouling mechanism of polysulfone ultrafiltration membrane in the treatment of coal gasification wastewater

Membrane fouling has been investigated by using a polysulfone ultrafiltration membrane with the molecular weight cutoff of 20 kDa to treat crushed coal pressurized gasification wastewater. Under the conditions of different feed pressures, the permeate flux declines and rejection coefficients of pollutants referring to three parameters (total organic carbon (TOC), chroma and turbidity) were studied. The membrane fouling mechanism was simulated with three classical membrane fouling models. The membrane image and pollutants were analyzed by scanning electron microscopy and gas chromatography-mass spectrography (GC-MS). The results indicate that the permeate flux decreases with volume reduction factor before reaching a constant value. The rejection coefficients were also measured: f_TOC = 70.5%, f_C = 84.9% and f_T = 91%. Further analysis shows that the higher the feed pressure is, the sooner the permeate flux reaches constant value and the more sharply the permeate flux declines. Constant flux indicates a nonlinear growth with feed pressure (P_F): when P_F equals 1.2 bar, the mark for the critical flux, slight membrane fouling occurs; when P_F exceeds 1.2 bar, cake layer pollution aggravates. Also the rejection coefficients of global pollutant increases slightly with P_F, suggesting the possibility of cake compression when P_F exceeds 1.2 bar. Through regression analysis, the fouling of polysulfone ultrafiltration membrane could be fitted very well by cake filtration model. The membrane pollutants were identified as phthalate esters and long-chain alkenes by GC-MS, and a certain amount of inorganic pollutants by X-ray photoelectron spectroscopy.

Open image in new window

     

An assessment of the Silt Density Index based on RO membrane colloidal fouling experiments with iron oxide particles

RO membrane colloidal fouling experiments were performed in the laboratory under well controlled and realistic conditions. Iron oxide was selected as a typical inorganic colloidal foulant, due to its importance, as evidenced from well known manufacturer recommendations on iron concentrations in feed waters and from frequently encountered problems in membrane installations. A range of iron concentrations was identified where a linear relationship existed between flux reduction rate and concentration. The performance of the Silt Density Index (SDI) was tested on the basis of the RO fouling data obtained. The range of iron concentrations where measurable and meaningful SDI values could be obtained was remarkably close to membrane manufacturer recommendations. A notable sensitivity of the SDI was also observed with particles for which retention is negligible. However, on the basis of the RO fouling data obtained, it appears that the SDI is not conservative enough. Furthermore, since the SDI cannot predict fouling rates, it cannot discriminate between different types of membranes.     

Evaluation of the fouling potential of ceramic membrane configurations designed for the treatment of oilfield produced water

During the treatment of oilfield produced water (PW) with ceramic membranes, process efficiency is primarily characterized by the specific permeate flux and the oil separation performance. In addition to membrane properties, the increase in total filtration resistance (fouling) and the decline in permeate flux are strongly dependent on the constituents of the PW as well as process conditions such as trans-membrane pressure and cross-flow velocity. The extensive study presented herein describes the characterization, application and performance of various ceramic membrane filtration technologies designed and developed for the efficient treatment of PW generated from tank dewatering and several oily model systems.     

Three‐dimensional simulation of the time‐dependent fluid flow and fouling behavior in an industrial hollow fiber membrane module

A novel three‐dimensional CFD model has been developed on the basis of fluid flow in the shell and lumen sides, and permeation and fouling behavior in the porous membrane zone. The simulated 25‐min dead‐end outside‐in filtration process showed that the energy consumed by the inlet manifold decreases during the constant pressure filtration. The velocity and pressure distributions in the module change with time. Flux distribution both in the axial and radial directions becomes increasingly more uniform, so does the cake distribution. Flux distribution and cake distribution inter‐adjust each other in different modes. A correlation equation has been developed to describe the relationship between the volumetric flow rate and accumulated water production. The correlation equation with simple experiment enables the dynamic evolution of energy consumed by shell inlet manifold to be presented, which can be the criterion of how well the shell inlet manifold or module has been designed. © 2018 American Institute of Chemical Engineers AIChE J, 64: 2655–2669, 2018